cardiogel supports adhesion, proliferation and ... · recently, cell therapy has emerged for the...

107 www.ecmjournal.org

P Sreejit et al. Cardiogel, a therapeutic biomaterialEuropean Cells and Materials Vol. 21 2011 (pages 107-121) DOI: 10.22203/eCM.v021a09 ISSN 1473-2262

Abstract

Cultured murine bone marrow derived mesenchymal stemcells (BMSC) when grown along with cardiogel derivedfrom mouse cardiac fibroblast, exhibited increased cellproliferation and differentiation and enhanced survivalunder oxidative stress induced by the exposure of H2O2 invitro (similar to in vivo ischemia like condition). Adhesionof BMSC to the cardiogel occurred at a faster rate whencompared to the cells grown on normal surface. BMSCattached to cardiogel showed an increased resistance toproteolytic (enzymatic) disassociation. This is the firstreport on an attempt to use an in house biomaterial for thegrowth of BMSC that led to their heightened resistancetowards oxidative stress. These studies support thatcardiogel is an efficient biodegradable three-dimensionalextracellular matrix which supports better growth of BMSCand can be used as a scaffold for stem cell delivery, withpotential therapeutic applications in cardiac tissueregeneration.

Key Words: Stem cells, proliferation, differentiation,cardiogel, oxidative stress protection, therapeuticbiomaterial.

*Address for correspondence:Rama Shanker VermaStem Cell & Molecular Biology Laboratory201, Bhupat and Jyoti Mehta School of Biosciences,Department of Biotechnology, Indian Institute ofTechnology Madras,Chennai - 600 036 TN, India

Telephone numbers: +91-44-22574109; 22575109E-mail: [email protected]

Introduction

It is well known that progression of myocardial infarctiongradually leads to heart failure (Struthers, 2005). Currentheart failure treatment is largely focused on maintainingthe residual overburdened cardiomyocytes (Christman andLee, 2006; Chen et al., 2008; Abbate et al., 2009).Recently, cell therapy has emerged for the treatment forseveral heart diseases (Costanzo et al., 1995; Heng et al.,2004; van Laake et al., 2006; Zhu et al., 2009). Celltherapy is based on the principle of supplementing theinsufficient inherent repair mechanisms within diseasedheart by utilizing either embryonic and/or adult stem cellsbesides mesenchymal stem cells, hematopoietic stem cells,umbilical cord blood cells and skeletal satellite myoblasts(Orlic et al., 2002; Reffelmann and Kloner, 2003;Dimarakis et al., 2005; Bettiol et al., 2007; Tang andPhillips, 2007; Catharina et al., 2008).

Undifferentiated stem cells can differentiate intomultiple lineages after transplantation (Jon et al., 2001;Donald and Darwin, 2007). Transplantation of stem/progenitor cells into damaged cardiac muscles has beenknown to induce differentiation into scar tissue containingfibroblasts (Wang et al., 2001). Under such circumstances,very few transplanted stem cells differentiate intocardiomyocytes, thereby reducing the clinical efficacy ofstem cell transplantation therapy for myocardialregeneration and finally affecting the revival of heartfunction after myocardial infarction. Microenvironment/niches, the environment in which stem cells are found,can interact to regulate stem cell fate, direct non-committed stem cells towards specific lineages and playa role in proliferation (Dimarakis et al., 2006a; Dimarakiset al., 2006b). It has been shown that transplanted humanembryonic stem cells (hESC) precommitted tocardiomyogenic lineage by prior co-culture with END2cells (mouse endoderm-like cell line), can survive,integrate, mature and proliferate after intramyocardialinjection in immunodeficient mice (van Laake et al.,2008).

However, stem cell transplantation in cardiac therapyby means of the direct injection method is subject to theloss of intercellular communication, extracellular matrix(ECM), and cell numbers. But its efficacy can beimproved by the use of patches of synthetic compositesor biomaterials as carrier (Ishii et al., 2005; Piao et al.,2007). In ischemic heart, these materials enhance thebinding of injected cells to the substratum at or near thesite of injury (ischemia) without disrupting the cell-cellmicroenvironment and prevent their loss by extrusion(Memon et al., 2005; Christman and Lee, 2006; Chen etal., 2008). Attempts have been made to replace infarcted

CARDIOGEL SUPPORTS ADHESION, PROLIFERATION AND DIFFERENTIATIONOF STEM CELLS WITH INCREASED OXIDATIVE STRESS PROTECTION

P. Sreejit and R.S. Verma*

Stem Cell and Molecular Biology Laboratory, Department of Biotechnology,Indian Institute of Technology Madras, Chennai-600036, TN, India


P Sreejit et al. Cardiogel, a therapeutic biomaterial

cardiac tissue with tissue-engineered cardiac patches madeof biocompatible and bioabsorbable materials like purifiedECM molecules and heterogeneous mixtures of ECMcomponents (Heng et al., 2004). Co-delivery of cells withvarious matrices including collagen matrices, Gel Foamand Matrigel has been successfully carried out (Kutschkaet al., 2006; Laflamme et al., 2007). But their long termeffects, including immunogenic response in most caseshave not been studied well.

Cardiogel is a naturally occurring extra cellular matrixderived from in vitro cultured fibroblasts. It is composedof laminin, fibronectin, and interstitial collagens (Types Iand III) besides other proteins, proteoglycans, and growthfactors (VanWinkle et al., 1996). ECM componentsconsiderably influence the growth characteristics ofcardiomyocytes, development of physiological activitiessuch as spontaneous contractile activity and phenotypemorphological differentiation, suggesting synergesticeffect of ECM components (Bick et al., 1998; Song et al.,2007). Elementary cardiac differentiation mechanismshave been elucidated using animal models along with invitro models of cardiac development such as mouseembryonal carcinoma cells, mouse embryonic stem cellsand recently, human embryonic stem cells (Kraehenbuehlet al., 2008; Beqqali et al., 2009). The growth rate andadhesion of murine mesenchymal stem cells in cardiogelwere found to increase when co-cultured together (Changet al., 2007).

In coronary diseases, narrowing of heart arteries resultin ischemia where there is a decreased blood and oxygensupply to heart, which can ultimately lead to a heart attack.Reactive oxygen species (ROS), a major cause of injuryafter ischemia/reperfusion, may hinder the adhesion andspreading of MSCs (Angelos et al., 2006; Song et al.,2010). Tissue hypoxia mediates an increase in the ROSburst at reperfusion and is associated with furtherimpairment of LV functional recovery. Efficacy of stemcell therapy using cardiogel in ischemic conditions hasnot been well studied. In the present study, we have culturedbone marrow derived stem cells isolated from adult micealong with cardiogel and report the survival of stem cellsagainst oxidative stress, which can occur duringreperfusion of ischemic areas of heart. The proliferativepotential of the bone marrow derived stem cells growingon cardiogel as well as the adhesion capability of stemcells was also evaluated.

Materials and Methods

Isolation and culture of bone marrow derivedmesenchymal stem Cells (BMSC)6-8 week-old Swiss albino mice were procured from TamilNadu Veterinary and Animal Sciences University, Chennai.All procedures, approved by the Institutional Animal EthicsCommittee (IIT Madras, India) and the Committee for thePurpose of Control and Supervision of Experiments onAnimals, Government of India, were performed under theRule 5(a) of the Breeding and Experiments on Animals(Control and Supervision) Rules (1998). All the animalsused for the experiments were killed by cervical

dislocation. Bone marrow-derived mesenchymal stem cells(BMSC) were collected from the aspirates of the femursand tibias of mice (~20 g) with 10 ml of BMSCmaintenance medium (BMM) consisting of DMEM/F12(1:1) medium (Invitrogen, Carlsbad, CA, USA)supplemented with fetal calf serum (FCS) (20%) (Gibco;Invitrogen), penicillin (100 U/ml), and streptomycin (100mg/ml; Invitrogen), 2 mM L-glutamine (Invitrogen), 0.1mM nonessential amino acids (Invitrogen) and 3 mMsodium pyruvate (Invitrogen). Cells were washed twiceand resuspended in BMM and plated at a density of 1 ×104 cells per cm2 in flasks. Cultures were maintained at37°C in a humidified atmosphere containing 5% CO2. After72 h, non-adherent cells were discarded, and the adherentcells were thoroughly washed twice with Dulbecco’sPhosphate Buffered Saline (DPBS). Fresh completemedium was added and replaced every 3 or 4 days for ~10days. The cells were harvested after incubation with 0.25%trypsin and 1 mM EDTA (Invitrogen) for 5 min at 37°Cand re-plated at densities of 5 × 103 cells per cm2. Thesecells were maintained for further studies and in mostexperiments cells between Passage Number (Pn) 5 and Pn12 were used. For experiments, plates coated with cardiogeland controls precoated with 0.2% gelatin were used. Theisolated cells were characterized by immunocytochemicalanalysis using antibodies for the surface markers CD29,CD34, CD45, (Santa Cruz Biotechnology, Santa Cruz, CA,USA), CD106 and Sca-1 (BioLegend, San Diego, CA,USA) with 1:100 dilution for primary antibodies and 1:200dilution for fluorophore conjugated secondary antibodies(Sigma-Aldrich, St. Louis, MO, USA) as described earlier(Chang et al., 2007).

Preparation of Cardiogel matrix coated platesCardiogel was prepared from confluent cardiac fibroblastsin-house using a modified version of a previously describedprotocol (Song et al., 2007). Briefly, the fibroblasts wereisolated by differential attachment of digested mousecardiac tissue suspension as described earlier (Sreejit etal., 2008). Whole hearts of 6-8 week-old Swiss albino micekilled by cervical dislocation, were excised andimmediately transferred into ice cold DPBS. Excised heartswere washed with chilled PBS, followed by sterile ice-cold balanced salt solution (20 mM HEPES-NaOH (pH7.6), 130 mM NaCl, 1 mM NaH2PO4, 4 mM glucose, 3mM KCl), in which the tissue was kept for 10 min. Thetissues were then minced with a sterile scalpel blade intosmall pieces ≤1 mm3 in 0.05% trypsin EDTA (Invitrogen)(0.3 ml per heart) and transferred into sterile 15 ml falcontubes. The myocardial cells were dispersed by incubatingwith 0.5% Trypsin EDTA (~1 ml of Trypsin for every 100mg of tissue) and were then mixed by intermittent pipettingalong with stirring at 37°C in a water bath for 4 min. Thecell suspension was allowed to stand for 1 min. Thesupernatant containing single cells was collected into 15ml falcon tube kept on ice, to which 2 ml DMEM/F12(1:1) medium (Invitrogen) supplemented with 20% fetalcalf serum (Invitrogen) was added and the digestion stepwas repeated thrice. The cell suspensions from eachdigestion were pooled and centrifuged at 2500g for 10min at 4°C.



The cell pellet was resuspended in a CardiacMaintenance Medium (CMM) containing DMEM/F12(1:1) medium supplemented with fetal calf serum (20%),horse serum (Invitrogen) (5%), penicillin (100 U/ml) andstreptomycin (100 mg/ml) (Invitrogen), 2mM L-glutamine(Invitrogen), 0.1mM non essential amino acids(Invitrogen), 3 mM sodium pyruvate (Invitrogen) andbovine insulin (1 μg/ml) (USV, Mumbai, India). Viabilityof cardiomyocytes was assessed by the Trypan blueexclusion test. The cells were plated on plates pre-coatedwith 1% gelatin and incubated in 95% air and 5% CO2 at37°C for ~2-3 h, to allow the differential attachment ofcardiac fibroblasts. The non-adherent cells were removedand the plates were washed twice with DPBS. The adherentcells were maintained in CMM. On attaining confluency,the adherent cardiac fibroblasts were trypsinized and re-plated at a density of 1×104 cells per cm2 into plates pre-coated with 0.2% gelatin. When these cells reachedconfluency, (3-4 days), the medium was carefully aspiratedand the plates were rinsed gently with PBS.

1 ml of pre-warmed extraction buffer (0.5% Triton X-100, 20 mM NH4OH in DPBS) was added to the plate,and the process of cell lysis was observed using an invertedmicroscope (Nikon Eclipse TS100, Melville, NY, USA),until no intact cells were visible (20-60 s). The cellulardebris were washed off with chilled DPBS and the plateswere stored at 37°C in DPBS.

Control plates and dishes were coated with 0.2% gelatindissolved in water. Gelatin suspension was added to thewells and coated by exposing it to ultraviolet (UV) rays of~62μW/cm2 intensity for 20 min to enhance cross-linking.The coated plates were incubated at 37°C for 2 h afterremoval of excess gelatin suspension and used for furtherexperiments.

Proliferation assayTo determine the rate of cell proliferation (fold increase)of BMSC in cardiogel, BMSC were seeded on 0.2% gelatincoated (control) and cardiogel coated plates at an initialdensity of ~5 x 103 cells per cm2. Cells were incubated for12 days at 37°C in a humidified 5% CO2 incubator. Cellgrowth at different time points (6th, 9th and 12th day) wasestimated by the MTT (3-(4,5-Dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. The above protocolwas done in conjunction with CFU-F assay, both of whichwere modifications of the protocols described earlier (Lapiet al., 2008).

MTT assayCell viability of the plated cells in cardiogel was measuredby the MTT assay (Song et al., 2007). The MTT assaywas used for cell proliferation studies as it measures thecellular activity of mitochondrial dehydrogenases. Theoptical density of the formazan product formed wasmeasured. The MTT assay was also used for understandingthe effects of stress in cells, since stress can influence thenumber of viable cells. Similarly in adhesion studies, thenumber of viable adherent cells in the wells remaining afterthe studies was estimated using the amount of MTT-formazan produced during the assay. Normal growth

medium was used as blank and appropriate cell controlsfor each assay were also kept for validating the obtaineddata. The experiments were performed thrice in duplicates.

Colony forming unit-fibroblast (CFU-F) assayThe MSC proliferative induction capability of cardiogelwas evaluated by CFU-F analysis. Bone marrow cells(passage number 2, 7 and 9) were seeded into the plates atan initial density of ~5 x 103 cells per cm2 in BMM. Cellswere incubated for 14 days at 37°C in a humidified 5%CO2 incubator. The medium was changed twice per week.On day 13, cultures were simultaneously fixed and stainedwith 0.1% crystal violet in 20% methanol by incubatingthem at room temperature for 30 min, and then washedtwice with PBS. Crystal violet stained fibroblastic colonieswith more than 50 cells were counted, to rule out colonyformation due to aggregation, and then photographed(Nikon Eclipse Ti). The experiment was performed intriplicate.

Stress induction studiesBMSC (initial inoculation of ~5 × 103 cells per cm2) weregrown on control and cardiogel coated plates. Cells (~60%confluency) were treated with various concentrations ofhydrogen peroxide (H2O2) (Merck, Darmstadt, Germany)in maintenance media for various time periods. H2O2 wasdiluted in Milli-Q water from the stock solution. Mediawas replenished and diluted H2O2 of various concentrationswas added into the well to obtain final concentrations of50 μM, 500 μM, 5 mM and 50 mM. After H2O2 addition,cells were incubated in dark at 37°C in a humidifiedatmosphere containing 5% CO2 for 1 to 4 days. After therequired exposure, the treatment media was decanted andfresh media was added. The MTT assay was carried out todetermine the cell growth and survival.

Cell adhesion studiesTo determine whether cell proliferation on cardiogel wasan outcome of cellular attachment to the scaffold, cellularadhesion assays were performed. The adhesion potentialof BMSC in cardiogel within a short time period wasassessed using a modified version of a previously describedprotocol (Vohra et al., 2008). BMSC were seeded ontocardiogel containing plates in concentrations ranging from103 to 104 cells per well of a 6 well plate and allowed toadhere for 1 h at 37°C. Following 1 h adhesion, wellscontaining cells were washed thrice with DPBS to removenon-adherent cells. The adherent cells were visualized andphotographed under an inverted microscope with attachedcamera (Nikon Eclipse Ti). The cell adhesion index wasquantified by the MTT assay.

To determine the intensity of cellular adhesion ofBMSC to cardiogel, which is reliant on the time requiredby the cells to disassociate from the surface, the followingprotocol was developed.

BMSC were also seeded on cardiogel containing platesand allowed to grow. After attaining ~70% confluency,the wells were washed with DPBS and then trypsinizedwith 0.25% Trypsin (Invitrogen) for varying time periods.The wells were exposed to trypsin for 0.5, 1 and 5 min



before neutralization with media containing 20% FCS. Thewells were washed thrice with DPBS to remove non-adherent cells. The adherent cells were visualized andimages recorded as described earlier. The cell adhesionindex was again measured by the MTT assay.

Scanning electron microscopyBMSC were seeded on glass cover slips coated withcardiogel and allowed to grow. After attaining ~70%confluency, the cover slips were chemically fixed with2.5% glutaraldehyde at room temperature and then seriallydehydrated in ethanol. The surface of the cover slip wassputter coated in a vacuum with an electrically conductive5 nm thick layer of Gold/Palladium alloy using a PrecisionEtching Coating system (Gatan, Warrendale, PA, USA;Model 682). Scanning electron microscopy (SEM) imageswere then recorded with a High Resolution SEM (FEI,Hillsboro, OR, USA; Quanta 200).

Statistical analysisStatistical analysis was performed by analysis of variance(ANOVA) and the Holm-Sidak test using the Sigma Plotversion 10 software package on the data obtained fromthe MTT assays conducted during cell proliferation studies,CFU-F assay and Stress Induction studies. P-values of≤0.05 indicate significant differences.

Results

Isolation and characteristics of BMSCBMSC were isolated from mixed cultures based on theirattachment on the culture plate. Isolated BMSC grew inpatches of long and slender cells, resembling the

characteristic morphology of bone marrow derivedmesenchymal stem cells. Initial passages showed thepresence of cell types with various other morphologicalfeatures. Most of these cells were macrophages, dendriticcells or other non stem cell populations found in the bonemarrow (Inaba et al., 1992). With the increase in passagenumber, gradual decline in non stem cells was observed.An increase in number of differentiated/progenitor cellswere also observed simultaneously. Hence in our study,cells of Pn 4 and above were used. When the isolated bonemarrow derived stem cells were characterized byimmunocytochemical analysis with antibodies against thefollowing markers: CD 29 (+ve); CD 106 (weak +ve); Sca– 1 (weak +ve); CD 45 (- ve); CD 34 (- ve), it was foundthat the cells obtained after Pn 2 were mostly MesenchymalStem Cells (MSCs) (data not shown).

Proliferation assay and CFU-F assayThe proliferation rate of BMSC was found to be higher incardiogel-coated plates than in normal plates (0.2% gelatincoated). Initially, noteworthy variations between theexpansion rates of BMSC grown in normal and cardiogelwere observed (one week). Two-Way ANOVA showedstatistically significant difference between the mean ofBMSC proliferation values (P = 0.032) among the differentdays (Day 3 and 7) after allowing for effects of differencesin surfaces (Cardiogel coated and Control). Similarly,statistically significant difference between the mean ofBMSC proliferation values (P = 0.049) among the differentsurfaces (Cardiogel coated and Control) was greater thanwould be expected after allowing the effects of differencesin days (Day 3 and 7) (Fig. 1). However, after 10 days thedifference gradually subsided as the cells were approachingconfluency (data not shown).

Fig. 1. Proliferative capacity of BMSCgrown on control and Cardiogel coatedplates. Graphical representation ofBMSC proliferation comparison incontrol and cardiogel coated platesusing MTT assay (initial density of ~5x 103 cells per cm2. Statisticalsignificance was calculated using 2-way ANOVA and compared withHolm-Sidak method (significancelevel = 0.05) – comparison between3rd day and 7th day (*P=0.032);comparison between Control andCardiogel plates (**P=0.049).



Fig. 2. CFU-F assay. (a) Graphical representation of the comparison of CFU-F colonies growing on control andcardiogel coated plates from different passages of BMSC (BMSC – initial density of ~5 x 103 cells per cm2).Statistical significance was calculated using 2-way ANOVA and compared with Holm-Sidak method (significancelevel = 0.05) – comparison between Pn 2, Pn 7 and Pn 9 (**P=<0.001); comparison between Control and Cardiogel(*P=0.002); (b,c,d) Images of BMSC (Passage number 2, 7 and 9 respectively) grown on control (0.2% Gelatincoated) plate showing scale bar of 100 μm size; (e,f,g) Images of BMSC (Passage number 2, 7 and 9 respectively)grown on cardiogel coated plates showing scale bar of 100 μm size.



The proliferative induction capability of cardiogel inBMSC to CFU-F was determined. To assess the increasein number of cells, MTT assay was performed to estimatethe number of viable cells. Low density seeding (~5 x 103

cells per cm2) was performed to attain optimal results andto rule out any deviation, cells were plated simultaneouslyand stained. The proliferative capability of BMSC to formCFU-F was found to be higher in cardiogel coated plates.A significant difference was observed in the number ofstained CFU-F colonies (with more than 50 cells) formedin normal and cardiogel coated plates (Fig. 2A). An overallincrease in number of stained CFU-F colonies wasobserved in higher passage number. With increase inpassage number, the difference between the number ofcolonies within the same passage numbers grown onnormal and cardiogel coated plates gradually decreased(data not shown). A difference of ~50% was observed incells in Pn 2. About 25 % difference was noticed in Pn 7and Pn 9. 2-way ANOVA showed statistically significant

difference between the mean values of number of CFU-Fcolonies formed (P = 0.002) at different passage numbers(Pn 2, 7, and 9) after allowing for the effects of differencesin surfaces (Cardiogel coated and Control). Similarlystatistically significant difference between the mean valuesof number of CFU-F colonies formed (P=0.011) ondifferent surfaces (Cardiogel coated and Control) wasgreater than that expected, after allowing for the effects ofdifferences in passage numbers (Pn 2, 7, and 9).

Cardiogel protective properties against stressOxidative stress was induced in BMSC grown on controland cardiogel coated plates and the protective effect ofcardiogel was assessed by considering the growth andviability of cells, as revealed by MTT assay. An increasein cell survivability and growth was observed in cellsgrown on cardiogel-coated plates as compared to the cellsgrown without cardiogel at higher concentrations of H2O2.

Fig. 3. Cell Viability assay of BMSCafter oxidative stress induction. Cellviability assay of BMSC in Controland Cardiogel coated Plates (postH2O2 treatment), with percentageviability calculated by consideringOD values of untreated wells as100%. Statistical significance wascalculated using 3-way ANOVA andcompared with Holm-Sidak method(significance level = 0.05) –comparison between 4 day and 1 dayH2O2 exposure (P=0.026);comparison between H2O2concentration 0, 0.5, 5 and 50 mM(**P=<0.001); comparison betweenControl and Cardiogel (*P=0.047).(a) After 1 day H2O2 treatment ; (b)After 4 day H2O2 treatment.



70% confluent cells were used in plates exposed toH2O2 for 1 day to allow the control wells to grow andbecome confluent. 60% confluent cells were used in platesexposed to H2O2 for 4 days to prevent the control wellsfrom overgrowing. This was done to avoid themisinterpretation of cellular activity and even cell deathin over grown cells as decrease in cell viability. Westandardized the post-plating out time for 60% and 70%confluent cells to be used for the stress induction studiesas 4 and 2 days respectively.

On exposing BMSC (~70% confluent) for 1 day withH2O2, less cell death was observed at lower concentrationsof H2O2. It was observed that after exposing cells to H2O2

concentration between 0.05 mM and 0.5 mM, nosignificant difference was observed in total cell growth,and cell death in both normal and cardiogel-coated.However, at higher concentration (5 mM) the percentageviability of BMSC in normal plates was ~10% of theviability in untreated plates where as the viability of BMSCin cardiogel-coated plates at similar condition was ~40%of the viability in untreated plates, which accounted 3 foldhigher than in normal plates (Fig. 3A).

Similarly, on exposing BMSC (~60% confluent) foradditional time (4 days) with H2O2, significant cell deathwas not observed at lower concentrations (0.05 mM) ofH2O2. But at higher concentration (0.5 mM), the percentage

Fig. 4. BMSC adhesion after 1 hour. (a,b,c) Images of BMSC growing on control (0.2% Gelatin coated) plateseeded at densities of 1, 5 and 10 X 103 cells per well of a 6 well plate showing scale bar of 50 μm size; (d,e,f) Imagesof BMSC growing on cardiogel coated plates seeded at densities of 1, 5 and 10 X 103 cells per well of a 6 well plateshowing scale bar of 50 μm size; (g) Graphical representation of cell viability assay on adherent BMSC growing oncontrol and cardiogel coated plates (after 1 h of adhesion).



Figure 5. Adhesion studies following enzymatic treatment. (a,b) Images of BMSC growing on control (0.2%gelatin coated) and cardiogel coated plate respectively without treatment used as Controls showing scale bar of 50μm size; (c,e) Images of BMSC growing on control (0.2% gelatin coated) plate after trypsinisation for 30 sec and1 min respectively showing scale bar of 50 μm size; (d,f) Images of BMSC growing on cardiogel plates aftertrypsinisation for 30 sec and 1 min respectively showing scale bar of 50 μm size; (g) Graphical representation ofcell viability assay of adherent BMSC growing on control and cardiogel coated plates after Trypsin treatment, withpercentage viability calculated by considering OD values of both the controls as 100%.



of viable BMSC in normal plates was ~70% as comparedto cardiogel-coated plates where cell count was ~80%. Thisdifference in viability increased up to 5 folds at higherH2O2 concentration (Fig. 3B). Three way analysis ofvariance (3-way ANOVA) was done with the above data,after the data from 0.05mM concentration was excludedfrom analysis. The results showed statistically significantdifference between the mean values of number of viablecells (P = 0.026) on different days of H2O2 exposure (1and 4 days) after allowing for effects of differences in H2O2concentrations (0.05, 0.5, 5 and 50 mM) and surfaces(Cardiogel coated and Control). Similarly statisticallysignificant difference between the mean values of numberof viable cells (P ≤0.001) at different H2O2 concentrations(0.05, 0.5, 5 and 50 mM) was greater than would beexpected after allowing the effects of differences in daysof H2O2 exposure (1 and 4 days) and surfaces (Cardiogelcoated and Control). Likewise, statistically significantdifference between the mean values of number of viablecells (P=0.047) among the different levels of surfaces(Cardiogel coated and Control) was found to be greaterthan expected after allowing the effects of differences indays of H2O2 exposure (1 and 4 days) and H2O2concentrations (0.05, 0.5, 5, and 50 mM).

Cell adhesion studiesTo demonstrate the adhesion intensity of cells in to thescaffold, two different assays were performed. For bothassays, 0.2% gelatin coated plates were used as normal(control). Firstly, to determine whether increased cellproliferation on cardiogel was an outcome of cellularattachment to the scaffold, cellular adhesion assays wereperformed. BMSC seeded at various concentrations ontonormal and cardiogel coated plates were allowed to adherefor 1 h at 37°C and the total number of viable adherentcells was then determined by MTT assay. The increasedOD during MTT assay in cardiogel coated plates correlated

with the higher number of adherent cells, visualized bymicroscopic observation (Nikon Eclipse Ti) (Fig. 4).

The second adhesiveness assay was performed toexamine the intensity of adhesion between the scaffoldand the mesenchymal stem cells. The intensity of cellularadhesion, a factor dependent on the cellular componentsand the molecules present on matrix surface, was correlatedto the time required by the cells to disassociate from thesurface. ~70% confluent BMSC grown on normal andcardiogel coated plates were trypsinized for varying timeperiods to remove the non-adherent cells and the extent ofcell adhesion was quantified by MTT assay. A conspicuousdifference in development of MTT was observed muchsimilar to that of the previous experiment. The percentageviability was calculated by considering both the controlsas 100%. The increased levels of MTT values in cardiogelcoated plates correlated with the presence of higher numberof adherent cells (Fig. 5).

SEM analysisSEM showed the level of attachment of BMSC oncardiogel. The cells were found to be enmeshed incardiogel (Fig. 6A) and the cellular appendages were foundto penetrate in to the cardiogel (Fig. 6B).

Discussion

In the present study, we elucidated the growthcharacteristics of bone marrow derived stem cell incardiogel. Our study demonstrated that BMSC were ableto survive, proliferate and adhere better in presence ofcardiogel than in normal environment during conditionsof stress. Comparative cell proliferation assay studies onnormal and cardiogel-coated plates confirmed an increasedcell growth. This could be attributed to augmented cellattachment to the scaffold. The results obtained from the

Fig. 6. SEM Images of BMSC grown on cardiogel coated surfaces showing enmeshment. (a) Normal view showingscale bar of 100 μm size; (b) Tilted view after gold palladium coating showing scale bar of 10 μm size.



modified 1 h adhesion assay evidently demonstrated anincreased cellular adhesion within a short time period incardiogel coated plates when compared to normal growthcondition (Lapi et al., 2008; Vohra et al., 2008) (Fig. 4).An alternate explanation for the increased cell growth incardiogel could be the combinatorial effect of numerousextracellular matrix components and signals in cardiogelthat would have stimulated faster cell division (Perkinsand Fleischman, 1990; VanWinkle et al., 1996). However,FACS analysis could not be performed to validate any ofthe above data sets as the cells were tightly bound to thescaffold and hence couldnot be disloged readily fromcardiogel. The adhesion assay using trypsinisation clearlyshows that ~40% of cardiogel adherent cells could bedisloged even after exposure to 0.25% trypsin for 1 min.Scraping of the cells for analysis could not be performedsince the scaffold tended to dislodge along with the cellsand enmesh around them. Hence it was not feasable toperform FACS analysis. This unique property shown bycardiogel of having a strong cellular adhesion to BMSCin a short time period could be useful in tissue engineeringapplications especially in case of delivery of stem cells into injured tissue. Injured tissue can be patched withcardiogel, which can then allow the binding of stem cellsdelivered directly to the site or intra venously and sinceonly a short time period is required, flow off of thedelivered stem cells to redundant sites can be prevented.This property also corroborates the previously describedusages of stem cell-seeded scaffold as an alternativetherapeutic modality for medically intractable advancedheart failure (Piao et al., 2007). The therapeuticapplications of cardiogel using stem cells has not beenreported, however many other scaffolds have been usedin conjunction with stem cells ( Christman and Lee, 2006;Chen et al., 2008).

As reported earlier, cardiogel can help in differentiationof embryonic stem cells into functional cardiomyocytes(Baharvand et al., 2005). The CFU-F assay results showedan augmentation of CFU-F colonies when cultured oncardiogel due to increase in cell differentiation. It wasobserved in the present study that the number of CFU-Fcolonies formed was much higher with increasing passagenumber. This increase in CFU-F colonies contradicted withdata obtained from previous reports (Perkins andFleischman, 1990). After a few passages of marrow cells,the predominant population was of CFU-F origin, whichin turn added up to the existing population (Fig. 2A). Ithas been already established that stromal cell progeny ofmurine bone marrow fibroblast colony-forming units areclonal endothelial-like cells, that express collagen IV andlaminin (Perkins and Fleischman, 1990). Production ofadhesion molecules by the cells and their progeny couldhave led to rapid proliferation of CFU-F cells. However,this could be also due to the usage of nutrient rich mediumcontaining 20% FCS for culture in the present study whichwould have resulted in the rapid expansion of CFU-Fprecursors (Biagioli et al., 2000).

A significant observation was made about the abilityof cardiogel to protect BMSC from stress. In the presentstudy, we used H2O2 to induce oxidative stress to BMSC.

Ischemia induces oxidative stress in tissues throughaccumulation of toxic substances especially due to freeradicals (Chen et al., 2000; Chang et al., 2006). Stressinduction using H2O2 exposure in vitro generates anischemia like condition much similar to that in vivo (Songet al., 2010). This condition was used by us to show theprotection offered by cardiogel. As compared to short termexposure (1-5 min) of H2O2 on BMSCs, better protectionresults were observed in long term exposure (1-4 days). Itis known that short term exposure with low-dose H2O2increases VEGF expression and enhances the survival andefficacy of BMSC for neovascularization and thus can beused in therapeutic angiogenesis (Kubo et al., 2007).Nevertheless, higher doses and long term exposure of H2O2increases cellular reactive oxygen species level, whichultimately leads to cell death. However, in our studysignificant reduction in cell death was observed, whenBMSC were grown on cardiogel. This could be attributedto the presence of extra-cellular matrix components, whichprovides sites for cellular binding and may also containgrowth factors (VanWinkle et al., 1996). The 3-dimensional architecture of the scaffold may also influencethe growth of BMSC in presence of H2O2 (Davis et al.,2006).

To summarize, our study demonstrated an increase incell proliferation and differentiation when BMSC weregrown on cardiogel. BMSC growing on cardiogel alsoshowed better oxidative stress tolerance induced by H2O2.Adhesion of BMSC to the scaffold occurred at a fasterrate when compared to control. Cells once adherent showedincreased resistance to dislodgment even by trypsinisationas clearly indicated in adhesion studies. Previous reportshad suggested the use of a biodegradable scaffold alongwith which, cells can be injected to the site of ischemia.The present study clearly demonstrated that BMSC oncegrown or allowed to adhere with cardiogel showed strongbinding affinity with scaffold. The cells proliferated,differentiated and even withstood the artificial oxidativestress. This when compared to other biomaterial scaffoldsincluding commercially available ones like Matrigel (BDBiosciences, Franklin Lakes, NJ, USA) could either delivercells or promote in situ regeneration (Kutschka et al., 2006;Laflamme et al., 2007). The long-term effects includingimmunogenic response of such materials are yet to becompletely understood. With this study, we havedemonstrated that cardiogel is an efficient potentiallybiodegradable three-dimensional extracellular matrix andcould be used for engineering cardiac tissue structures forcell culture and potential therapeutic applications postischemic cardiac tissue regeneration. Further work withanimal models will confirm the applicability of thisbiomaterial for therapeutic purposes.

Acknowledgments

This work was supported by grants to RSV by theDepartment of Biotechnology, Govt. of India (BT/PR7951/MED/14/1193/2006) and by fellowship to SP by the IndianCouncil for Medical Research (No. 3/1/3/JRF/37/MPD/2004(32402)).



References

Abbate A, Biondi-Zoccai GGL, Van Tassell BW, BaldiA (2009) Cellular preservation therapy in acute myocardialinfarction. Am J Physiol Heart Circ Physiol 296: H563-565.

Angelos MG, Kutala VK, Torres CA, He G, Stoner JD,Mohammad M, Kuppusamy P (2006) Hypoxic reperfusionof the ischemic heart and oxygen radical generation. Am JPhysiol Heart Circ Physiol 290: H341-347.

Baharvand H, Azarnia M, Parivar K, Ashtiani SK(2005) The effect of extracellular matrix on embryonicstem cell-derived cardiomyocytes. J Mol Cell Cardiol 38:495-503.

Beqqali A, van Eldik W, Mummery C, Passier R (2009)Human stem cells as a model for cardiac differentiationand disease. Cell Mol Life Sci 66: 800-813.

Bettiol E, Clement S, Krause KH, Jaconi ME (2007)Embryonic and adult stem cell-derived cardiomyocytes:Lessons from in vitro models. Rev Physiol BiochemPharmacol 157: 1-30.

Biagioli B, Maccherini M, Carlucci F, Tabucchi A,Matucci R, Mugelli A (2000) Oxidative stress: Clinicalimplications in cardiac surgery. Ital Heart J 1: S43-S45.

Bick RJ, Snuggs MB, Poindexter BJ, Buja LM, VanWinkle WB (1998) Physical, contractile and calciumhandling properties of neonatal cardiac myocytes culturedon different matrices. Cell Comm Adhes 6: 301-310.

Catharina N, Nan M, Karen B, Wolfgang W, AnthonyH, Yrjö TK, Hao Z, Mihail EH, Gustav S (2008)Mesenchymal stem cells and cardiac repair. J Cell MolMed 12: 1795-1810.

Chang W, Kim J-Y, Lim S, Lee S, Song B-W, Kim H-J, Cha M-J, Kwon S-Y, Han S-M, Min B-H, Jang Y, ChungN, Hwang K-C (2006) Mesenchymal stem cells withcalreticulin gene modulate cell adhesiveness through anintegrin-mediated mechanism. Tissue Eng Regen Med 3:327-335.

Chang W, Lim S, Song H, Lee S, Song B-W, Jang Y,Chung N, Hwang K-C (2007) In vitro expansion ofmesenchymal stem cells using 3-D matrix derived fromcardiac fibroblast. Tissue Eng Regen Med 4: 370-375.

Chen QM, Tu VC, Wu Y, Bahl JJ (2000) Hydrogenperoxide dose dependent induction of cell death orhypertrophy in cardiomyocytes. Arch Biochem Biophys373: 242-248.

Chen Q-Z, Harding SE, Ali NN, Lyon AR, BoccacciniAR (2008) Biomaterials in cardiac tissue engineering: tenyears of research survey. Mater Sci Eng R Rep 59: 1-37.

Christman KL, Lee RJ (2006) Biomaterials for thetreatment of myocardial infarction. J Am Coll Cardiol. 48:907-913.

Costanzo MR, Augustine S, Bourge R, Bristow M,O’Connell JB, Driscoll D, Rose E (1995) Selection andtreatment of candidates for heart transplantation: Astatement for health professionals from the committee onheart failure and cardiac transplantation of the council onclinical cardiology, American Heart Association.Circulation 92: 3593-3612.

Davis RA, Van Winkle WB, Buja LM, Poindexter BJ,Bick RJ (2006) Effect of a simple versus a complex matrixon the polarity of cardiomyocytes in culture. J BurnsWounds 5: 16-25.

Dimarakis I, Habib NA, Gordon MYA (2005) Adultbone marrow-derived stem cells and the injured heart: Justthe beginning? Eur J Cardiothorac Surg 28: 665-676.

Dimarakis I, Levicar N, Nihoyannopoulos P, GordonMY, Habib NA (2006a) In vitro stem cell differentiationinto cardiomyocytes: Part 2: Chemicals, extracellularmatrix, physical stimuli and coculture assays. JCardiothorac Renal Res 1: 115-121.

Dimarakis I, Levicar N, Nihoyannopoulos P, HabibNA, Gordon MY (2006b) In vitro sem cell differentiationinto cardiomyocytes: Part 1. Culture medium and growthfactors. J Cardiothorac Renal Res 1: 107-114.

Donald GP, Darwin JP (2007) Concise review:Mesenchymal stem/multipotent stromal cells: The state oftransdifferentiation and modes of tissue repair – currentviews. Stem Cells 25: 2896-2902.

Heng BC, Haider HK, Sim EK-W, Cao T, Ng SC (2004)Strategies for directing the differentiation of stem cells intothe cardiomyogenic lineage in vitro. Cardiovasc Res 62:34-42.

Inaba K, Inaba M, Romani N, Aya H, Deguchi M,Ikehara S, Muramatsu S, Steinman RM (1992) Generationof large numbers of dendritic cells from mouse bonemarrow cultures supplemented with granulocyte/macrophage Colony-Stimulating Factor. J Exp Med 176:1693-1702.

Ishii O, Shin M, Sueda T, Vacanti JP (2005) In vitrotissue engineering of a cardiac graft using a degradablescaffold with an extracellular matrix-like topography. JThorac Cardiovasc Surg 130: 1358-1363.

Jon SO, Dan SK, James AT (2001) Multilineagedifferentiation from human embryonic stem cell lines. StemCells 19: 193-204.

Kraehenbuehl TP, Zammaretti P, Van der Vlies AJ,Schoenmakers RG, Lutolf MP, Jaconi ME, Hubbell JA(2008) Three-dimensional extracellular matrix-directedcardioprogenitor differentiation: Systematic modulation ofa synthetic cell-responsive peg-hydrogel. Biomaterials 29:2757-2766.

Kubo M, Li T-S, Suzuki R, Ohshima M, Qin S-L,Hamano K (2007) Short-term pretreatment with low-dosehydrogen peroxide enhances the efficacy of bone marrowcells for therapeutic angiogenesis. Am J Physiol Heart CircPhysiol 292: H2582-2588.

Kutschka I, Chen IY, Kofidis T, Arai T, von DegenfeldG, Sheikh AY, Hendry SL, Pearl J, Hoyt G, Sista R, YangPC, Blau HM, Gambhir SS, Robbins RC (2006) Collagenmatrices enhance survival of transplanted cardiomyoblastsand contribute to functional improvement of ischemic rathearts. Circulation 114: I-167-173.

Laflamme MA, Chen KY, Naumova AV, Muskheli V,Fugate JA, Dupras SK, Reinecke H, Xu C, HassanipourM, Police S, O’Sullivan C, Collins L, Chen Y, Minami E,Gill EA, Ueno S, Yuan C, Gold J, Murry CE (2007)Cardiomyocytes derived from human embryonic stem cells



in pro-survival factors enhance function of infarcted rathearts. Nat Biotechnol 25: 1015-1024.

Lapi S, Nocchi F, Lamanna R, Passeri S, Iorio M,Paolicchi A, Urciuoli P, Coli A, Abramo F, Miragliotta V,Giannessi E, Stornelli M, Vanacore R, Stampacchia G,Pisani G, Borghetti L, Scatena F (2008) Different mediaand supplements modulate the clonogenic and expansionproperties of rabbit bone marrow mesenchymal stem cells.BMC Res Notes 1: 53.

Memon IA, Sawa Y, Fukushima N, Matsumiya G,Miyagawa S, Taketani S, Sakakida SK, Kondoh H, AleshinAN, Shimizu T, Okano T, Matsuda H (2005) Repair ofimpaired myocardium by means of implantation ofengineered autologous myoblast sheets. J ThoracCardiovasc Surg 130: 1333-1341.

Orlic D, Hill JM, Arai AE (2002) Stem cells formyocardial regeneration. Circ Res 91: 1092-1102.

Perkins S, Fleischman RA (1990) Stromal cell progenyof murine bone marrow fibroblast colony-forming unitsare clonal endothelial-like cells that express collagen IVand laminin. Blood 75: 620-625.

Piao H, Kwon J-S, Piao S, Sohn J-H, Lee Y-S, Bae J-W, Hwang K-K, Kim D-W, Jeon O, Kim B-S, Park Y-B,Cho M-C (2007) Effects of cardiac patches engineeredwith bone marrow-derived mononuclear cells and PGCLscaffolds in a rat myocardial infarction model. Biomaterials28: 641-649.

Reffelmann T, Kloner RA (2003) Cellularcardiomyoplasty – Cardiomyocytes, skeletal myoblasts,or stem cells for regenerating myocardium and treatmentof heart failure? Cardiovasc Res 58: 358-368.

Song H, Chang W, Lim S, Seo H-S, Shim CY, Park S,Yoo K-J, Kim B-S, Min B-H, Lee H, Jang Y, Chung N,Hwang K-C (2007) Tissue transglutaminase is essentialfor integrin-mediated survival of bone marrow-derivedmesenchymal stem cells. Stem Cells 25: 1431-1438.

Song H, Cha M-J, Song B-W, Kim I-K, Chang W, LimS, Choi E, Ham O, Lee S-Y, Chung N, Jang Y, Hwang K-C (2010) Reactive oxygen species inhibit adhesion ofmesenchymal stem cells implanted into ischemicmyocardium via interference of focal adhesion complex.Stem Cells 28: 555-563.

Sreejit P, Kumar S, Verma RS (2008) An improvedprotocol for primary culture of cardiomyocyte fromneonatal mice. In Vitro Cell Dev Biol Anim 44: 45-50.

Struthers AD (2005) Pathophysiology of heart failurefollowing myocardialinfarction. Heart 91: 14-16.Tang YL, Phillips MI (2007) Invited commentary. AnnThorac Surg 83: 1499-1500.

Van Laake LW, Hassink R, Doevendans PA, MummeryC (2006) Heart repair and stem cells. J Physiol 577: 467-478.

Van Laake LW, Passier R, Doevendans PA, MummeryCL (2008) Human embryonic stem cell-derivedcardiomyocytes and cardiac repair in rodents. Circ Res102: 1008-1010.

VanWinkle WB, Snuggs MB, Buja LM (1996)Cardiogel: A biosynthetic extracellular matrix forcardiomyocyte culture. In Vitro Cell Dev Biol Anim 32:478-485.

Vohra S, Hennessy K, Sawyer A, Zhuo Y, Bellis S(2008) Comparison of mesenchymal stem cell andosteosarcoma cell adhesion to hydroxyapatite. J Mater SciMater Med 19: 3567-3574.

Wang J-S, Shum-Tim D, Chedrawy E, Chiu RCJ (2001)The coronary delivery of marrow stromal cells formyocardial regeneration: Pathophysiologic and therapeuticimplications. J Thorac Cardiovasc Surg 122: 699-705.

Zhu W-Z, Hauch KD, Xu C, Laflamme MA (2009)Human embryonic stem cells and cardiac repair. TransplantRev (Orlando) 23: 53-68.

Discussion with Reviewers

Reviewer I: Is it known which specific ECM componentsinfluence growth characteristics? If so, please describe!Authors: Till date the ECM components present inCardiogel have not been completely characterized. We areattempting to characterize those components, in order todetermine the specific components which influence growthand differentiation characteristics.

Reviewer I: The description of the presence of cells withother morphologies indicates a mixed preparation – this isto be expected with such a culture though. Is it knownwhat happens to these cells: do they dedifferentiate, takeon stem cell characteristics, die, etc.? Althoughimmunocytochemical analysis demonstrates that cells aregenerally stem cells, it would be of interest to know whathappens to these other cells.Authors: We always obtained cells with othermorphologies clearly indicating a mixed culture duringthe initial passages. Most of these cells are macrophages,dendritic cells or other non stem cell populations found inthe bone marrow. Almost all these cells are highlydifferentiated except for the progenitor cells of the abovepopulations and do not have the capacity to attain stemcell characteristics, unless newer technologies like iPS areused. Since these cells are mostly terminally differentiated,these cells cannot be maintained for more than 2-4 passagesunless specific growth factors are added to allow theproliferation of the progenitor cells. (Inaba et al., 1992,text reference).

Reviewer I: Regarding your stress data description: H2O2exposure can lead to apoptosis. It would be of interest tosee whether the described protection against H2O2-inducedstress by cardiogel is due to less apoptosis – it isrecommended that the authors compare H2O2-inducedapoptosis (e.g., DNA laddering, TUNEL assay, caspaseactivation, etc.) in cells plated out on gelatin and cardiogelcoated plates – additional controls could include comparingeffects to other scaffolds such as Matrigel – that wouldindicate whether protective effects are due specifically tocardiogel, or if they are as a result of general 3D cultureconditions.Authors: We agree with the point that the reviewer hasraised about the apoptosis of cells due to H2O2 exposure.The data of these experiments will surely point out the



mechanism of protection provided by cardiogel. But alongwith these experiments, other molecular studies and assaysneeds to be done to elucidate the mechanism of cellproliferation and adhesion in cardiogel, to give thecomplete picture of how the BMSC binds to the cardiogeland how it is protected against stress. Besides oxidativestress, other stress like mitochondrial stress and hypoxiahas to be also compared. We are already working in thisdirection.

Reviewer I: Why was the ‘0.05 mM’ data excluded fromthe analysis?Authors: The 0.05 mM data has been excluded from theanalysis since the data was considered as an outlier throughstatistical analysis. Secondly, previous studies with lowdose exposure of H2O2 have been shown to increase BMSCproliferation. However high dose exposure of H2O2 clearlyshow a decrease in cell viability which occurs due to stressinduced by H2O2 exposure. Since this assay was done tocompare the protection capability of cardiogel coated andcontrol plates to MBMC against oxidative stress inducedthrough H2O2 exposure, therefore the data points abovethe controls were not considered for statistical analysis.However high dose exposure of H2O2 clearly show adecrease in cell viability which occurs due to stress inducedby H2O2 exposure. Since this assay was done to comparethe protection capability of cardiogel coated and controlplates to MBMC against oxidative stress induced throughH2O2 exposure, therefore the data points above the controlswere not considered for statistical analysis.

Reviewer I: Why is the BrdU labeling data not given?This would be a better measure of proliferation than MTTdata; MTT measures the relative number of metabolicallyactive cells (this also encompasses cell death), whereasBrdU directly measures cells showing DNA synthesis, e.g.,actively dividing. It is recommended that BrdU data isgiven; both micrographs and cell count data (number ofBrdU positive cells/total cells – if it is not possible to dothis by FACS analysis, then cell counts ofimmunofluorescently labeled cells may be employed – seeEdwards et al. (2008) for method and analysis details.Authors: Our study shows that cells were tightly boundto the scaffold and hence could not be dislodged readilyfrom cardiogel. As a result BrdU labeling studies couldnot be performed since FACS analysis needs cellulardislodgement and prolonged trypsinisation to dislodge thecells may show erroneous results. We also accept the pointraised by the reviewer that BrdU labeling is a bettermeasure of proliferation than MTT, since BrdU directlymeasures cells showing DNA synthesis. An attempt wasmade to replicate the methodology suggested by thereviewer, but non-specific staining of scaffold with Antianti BrdU antibody in the background makes cell countingthrough BrdU labeling a challenge.

Reviewer I: Serum in the media can influence cellproliferation, protection against oxidative stress, etc. Itwould be of interest to see how the cells proliferate,differentiate and respond to oxidative stress on control and

cardiogel surfaces with lower serum concentrations or inthe absence of serum.Authors: We agree that it would be quite interesting tostudy the influence of serum on cell proliferation,differentiation and response to oxidative stress on controland cardiogel surfaces. Since our aim to replicate the invivo condition in vitro where the tissue is constantlysupplied by fluids like blood or lymph, this does not fallunder the main aim of the study.

Reviewer I: Is it known what the lifetime of H2O2 in cultureis; it may become expended rapidly. Were cells exposedto a single dose or, in experiments where cells were treatedfor 1-4 days, were the media containing H2O2 replacedwith fresh media containing H2O2 periodically?Authors: The lifetime of H2O2 in culture is not knownand it may be expended rapidly. However our data asshown in Fig. 3A and 3B proves that, even after prolongedtreatment for 4 days, H2O2 in final concentrations of 0.5and 5 mM were able to induce more stress compared tothat treated for 1 day. In our studies, the cells were exposedto only a single dose, even when cells were treated for 1-4 days and the media containing H2O2 was not at allreplaced with fresh media containing H2O2 periodically.

Reviewer I: A reduction in cell death may also be due tothe cardiogel/ECM or serum in the medium quenchingreactive oxygen species?Authors: We agree with the reviewer that reduction incell death may also be due to the cardiogel/ECM quenchingreactive oxygen species. However the point that serum inthe medium may quench reactive oxygen species isdebatable since, significant difference in cell viability wasobserved between cells grown on cardiogel coated andcontrol plates. Besides prolonged H2O2 treatment (finalconcentrations of 0.5 and 5 mM) for 4 days were able toinduce more stress compared to that treated for 1 day. Thusit is quite unlikely that serum may quench reactive oxygenspecies and thus influence cell viability.

Reviewer II: Although the authors tested the effects ofoxidative stress on marrow stromal cells grown oncardiogel, it is not clear how this mimics a clinical scenario.By the time such an implantation may occur in a host thisoxidative stress related condition may not prevail.Authors: Stem cell transplantation in cardiac therapy bymeans of the direct injection method is subject to the lossof intercellular communication, extracellular matrix, andcell numbers. Implantation of stem cell sheets might bemore advantageous in repairing the impaired heart byproviding uniform and stable cell delivery with less cellloss and without disrupting the cell-cell microenvironment(Memon et al., 2005, text reference). Bioengineeredcardiac grafts have been created previously by combiningautologous cell transplantation with a degradable scaffoldas a temporary extracellular matrix (Ishii et al., 2005, textreference). Recently, two studies from different groupshave reported on successful engraftment of cardiomyocytesis enhanced by co-delivery of the cells with matrix.Kutschka and co-workers have demonstrated that grafts



were larger and ventricular function was improved ininfarcted rat hearts, when human ESC-derivedcardiomyocytes were delivered together with collagenmatrices, including Gel Foam and the basement membranepreparation, Matrigel (Kutschka et al., 2006, textreference). Matrigel was found to significantly enhancesurvival of hESC-derived cardiomyocytes in infarcted rathearts (Laflamme et al., 2007, text reference). Reactiveoxygen species (ROS), a major cause of injury afterischemia/reperfusion, may hinder the adhesion andspreading of MSCs. Generation of ROS is dramaticallyincreased over three-fold in the risk region in ischemia/reperfusion-injured animal hearts compared to wild-typeanimal hearts (Song et al., 2010, text reference). Tissuehypoxia mediates an increase in the ROS burst atreperfusion and is associated with further impairment ofLV functional recovery. The greatest increase in ROSformation was noted in the most hypoxic myocardial tissueat the time of reperfusion (Angelos et al., 2006, textreference). Our study was directed to assess the survivaland proliferation potential of stem cells grown on cardiogelas cell support sheets and also to assess tolerance tooxidative stress in vitro.

We accept the point raised by the reviewer that by thetime such an implantation may occur in a host this oxidativestress related condition may not prevail. However therecurrence of cardiac infarction and increase in size ofischemic area remains a possibility due to the continuedpresence of diseased coronary arteries. The repair to theheart muscle is not always complete and scarring is usuallypresent. There is also the risk of heart failure developingover a period of weeks as the heart reacts to the injury ithas sustained. For these reasons it is necessary for patientsto be monitored carefully and to receive the appropriatetreatment to reduce the risk of further disease progressionand other heart attacks. Medical treatment is aimed atpreventing further damage and the chance of repeat heartattacks in the future by opening the blocked artery andrestoring the blood flow to the affected area of heart muscleby reperfusion. However reperfusion as previously saidincreases a three-fold generation of ROS.

Reviewer II: The authors demonstrated that stromal cellscultured on cardiogel are more adherent than cells culturedon control plates. The authors discussed about the possibleapplications of cardiogel as a scaffold to grow ordifferentiate cardiomyocytes from other cell sources butdid not discuss how such cardiogel containing grown cellswould be removed intact and implanted since they are veryadherent. Others have shown that oxidative stressdiminishes the adhesiveness of stromal cells on cardiogel(Song et al., 2010, text reference), this need to be discussedin this context. Also, could you describe the status ofstromal cells following H2O2 treatment?Authors: The reviewer has raised a valid questionregarding the removal and implantation of cardiogelcontaining grown cells. Cardiogel by nature is adherentand would be difficult to remove. However, during ourstudies we have found that cardiogel tends to get dislodgedfrom surface by altering the pH of the solution. Another

option is by using scrapers to remove a portion and to peeloff the remaining portion, but this is effective for thickerand larger gels. After the dislodgment, the cardiogel withcells can be directly implanted at the site of injury. In caseof delivery through injection, the cardiogel with cells maybe completely scrapped even without altering the pH ofmedia. It is however advisable to use the implantationmethod to avoid the loss of intercellular communication,extracellular matrix, and cell numbers.

Song et al. (2010, text reference) did demonstrate thatoxidative stress diminished the adhesiveness of stromalcells on cardiogel. But the authors also pointed out thatthe cell adhesions to fibronectin-coated plates and thecardiogel were significantly increased as compared withadhesion to bare plates. Only H2O2 treatment significantlyreduced the amount of adhered MSCs at the indicatedtimes, regardless of the coating material. Our datacorroborate well with the above findings that cell adhesionsto the cardiogel coated plates were significantly increasedas compared with adhesion to bare plates and also thatH2O2 treatment significantly reduced the amount ofadherent cells.

Reviewer II: An explanation regarding why a certainpassage number of marrow stromal cells was chosen wouldbe helpful. The authors describe increase cell survivabilityof cells cultured on cardiogel rather than reduced survivalof the cells in control culture.Authors: Most of the previous reports have used BMSCwithin the 15 passages of initial plating. At latter passagesdue to increase in spontaneous differentiation and increasedproliferation of differentiated cells, there is a net decreasein number of BMSC. Therefore, we have used BMSCbelow passage number 15 for our studies. Our study wasdirected towards utilizing cardiogel as a therapeuticscaffold. Therefore, we were interested in determining theincrease in survivability of cells grown on cardiogel andits comparison with cells grown on control plates.

Reviewer II: What is the mechanism of protection fromoxidative stress in bone marrow stromal cells cultured oncardiogel and why does this protection diminish with time?Authors: The mechanism of protection from oxidativestress in bone marrow stromal cells cultured on cardiogelhas not been completely elucidated. We suspect that thereduction in cell death may be due to the ROS quenchingby cardiogel/ECM. However, it should be also noted thatthe medium or its constituents may likely quench ROS isdebatable. Significant difference in cell viability was alsoobserved between cells grown on cardiogel coated andcontrol plates. Prolonged H2O2 treatment (finalconcentrations of 0.5 and 5 mM) for 4 days was able toinduce more stress compared to that treated for 1 day. Thusit is quite unlikely that media may quench ROS and thusinfluence cell viability. Therefore the only parameter whichmay quench the ROS in the experiment is the cardiogel/ECM. Similarly, by using N-acetyl L-cysteine (free radicalscavenger) (Song et al., 2010, text reference) for co-treatment in H2O2 treated MSCs, adhesion and spreadinginhibition was reduced and detachment from surface



decreased. Therefore the reduction in cell death might bedue to the cardiogel/ECM quenching ROS. Further studiesare required to prove this observation.

The diminished protection, which arose with time maybe due to the decreased quenching of ROS by cardiogel/ECM over the treatment period. Another possibility is dueto the activation of alternative pathways by the scavengingROS on long-term exposure.

Additional Reference

Edwards GO, Jondhale S, Chen T, Chipman JK (2008) Aquantitative inverse relationship between connexin32expression and cell proliferation in a rat hepatoma cellline. Toxicology 253: 46-52.

cardiogel supports adhesion, proliferation and ... · recently, cell therapy has emerged for the...

Documents